pith. sign in

arxiv: 2607.00090 · v1 · pith:OERH6QLSnew · submitted 2026-06-30 · 💻 cs.CV · cs.AI

Lost in the Tail: Addressing Geographic Imbalance in Urban Visual Place Recognition

Zhiyao Shu , Jiacheng Yang , Yang Lu , Waishan Qiu , Chuan Li , Da Chen This is my paper

Pith reviewed 2026-07-02 19:53 UTC · model grok-4.3

classification 💻 cs.CV cs.AI
keywords visual place recognitionlong-tailed distributiongeographic imbalanceurban datasetsgradient rebalancingmulti-scale searchplug-in framework
0
0 comments X

The pith

Rebalancing gradients across head and tail classes corrects geographic bias in urban visual place recognition.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper shows that urban visual place recognition datasets have a long-tailed distribution that causes models to perform well only on frequently photographed locations. It proposes a plug-in called Distribution-Aware Place Recognition that rebalances the training gradients to treat all locations more equally and adds a multi-scale distance search in the retrieval phase. These changes produce large accuracy gains on the SF-XL benchmark and work when added to other methods on several datasets. Readers should care because place recognition systems need to work in every part of a city, not just the popular ones, for applications like mapping and autonomous navigation.

Core claim

Urban-scale VPR suffers from long-tailed imbalance that biases models toward locations with abundant images. Distribution-Aware Place Recognition addresses this as a model-agnostic plug-in by rebalancing gradient contributions across head and tail classes and, within classification-retrieval pipelines, applying multi-scale distance search to compute per-class distributional compactness, which yields 18.3% improvement on SF-XL test set v1 and 6.7% on v2 while providing consistent gains across methods and benchmarks.

What carries the argument

Distribution-Aware Place Recognition (DAPR) as a model-agnostic plug-in framework that rebalances gradient contributions across head and tail classes and applies multi-scale distance search for per-class distributional compactness.

If this is right

  • Delivers 18.3% higher performance than the prior baseline on SF-XL test set v1.
  • Delivers 6.7% higher performance on SF-XL test set v2.
  • Improves multiple existing VPR methods when added as a plug-in module on SF-XL, MSLS and Pitts30k.
  • Reduces the systematic favoritism toward frequently photographed locations.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Methods like DAPR could help other computer vision tasks that face similar long-tailed geographic data distributions.
  • Real-world deployment in cities could extend reliable place recognition to less-visited areas.
  • Dataset creators might use the identified imbalance metrics to design more balanced collection strategies.

Load-bearing premise

Rebalancing gradients and multi-scale search will improve tail class performance without introducing new biases or requiring dataset-specific tuning.

What would settle it

An experiment showing that DAPR produces no gain over the baseline when tested on a new urban VPR dataset with a different long-tail structure would falsify the claim of broad generalizability.

Figures

Figures reproduced from arXiv: 2607.00090 by Chuan Li, Da Chen, Jiacheng Yang, Waishan Qiu, Yang Lu, Zhiyao Shu.

Figure 1
Figure 1. Figure 1: Geographic classes in SF-XL [9] are ranked by sample count (blue curve) and partitioned into head (top 30%), middle (40%), and tail (bottom 30%). Intra-class fea￾ture diversity (green band) indicates richer and more varied visual content in sparsely sampled regions (widens toward tail classes). Per-class Recall@1 (orange curve), mea￾sured on the D&C [48] model, declines sharply toward the tail. discriminat… view at source ↗
Figure 2
Figure 2. Figure 2: The geographic classes distribution of SF_XL [9] on real-world map. Red re￾gions (head classes) represent frequently photographed locations from SF_XL [9]. Or￾ange regions (middle classes) show moderately sampled areas, and green regions (tail classes) indicate sparsely photographed locations. Four representative regions (1-4) are zoom-in spatial patterns of class imbalance: region 1 and 4 reveal predomina… view at source ↗
Figure 3
Figure 3. Figure 3: Distribution-Aware Place Recognition Framework. [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figure 5
Figure 5. Figure 5: Comparison of Retrieval Dis￾tance on D&C Method with SF-XL test v1 and test v2. R@5 on test v1), as classification heads rely on per-class decision boundaries and are thus more sensitive to class imbalance than contrastive methods that inher￾ently emphasize hard examples. Combining Llb with the D&C mixed pipeline further adds +2.3% R@1, and gains are consistent across both test sets, confirm￾ing robustness… view at source ↗
Figure 6
Figure 6. Figure 6: Per-class Recall@1 across head, middle, and tail bins defined by leaf spatial extent (m2 ) on CPlaNet’s [42] of SF-XL (131,080 leaves). while the inference time grows more than 10 times due to their feature size re￾quirements. Moreover, SALAD and BoQ reshape DINOv2 features into cluster aggregators that the DAPR classifier cannot reuse, so the pipeline maintains two independent DINOv2 backbones at both the… view at source ↗
read the original abstract

Urban-scale Visual Place Recognition (VPR) aims to identify the geographic location of a query image by matching it against a geo-tagged database. While recent methods achieve impressive performance, they overlook a serious long-tailed problem hidden in urban-scale datasets, which biases the model towards locations with abundant images and ignores less-visited areas, causing models to systematically favor frequently photographed locations while failing in sparsely covered areas. In this paper, we systematically characterize this imbalance challenge and propose Distribution-Aware Place Recognition (DAPR), a model-agnostic plug-in framework that rebalances gradient contributions across head and tail classes. Additionally, within classification-retrieval pipelines, DAPR applies a multi-scale distance search mechanism to compute per-class distributional compactness, providing complementary gains at the retrieval stage. On the large-scale SF-XL benchmark, our framework outperforms the previous classification-retrieval baseline by 18.3% on test set v1, and 6.7% on test set v2. As a plug-in module, it achieves consistent improvements across representative VPR methods on SF-XL, MSLS, and Pitts30k, demonstrating broad generalizability across different methods and benchmarks.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 0 minor

Summary. The paper identifies a long-tailed geographic imbalance in urban visual place recognition (VPR) datasets that biases models toward frequently photographed locations. It proposes Distribution-Aware Place Recognition (DAPR), a model-agnostic plug-in framework that rebalances gradient contributions across head and tail classes; within classification-retrieval pipelines it additionally applies multi-scale distance search to compute per-class distributional compactness. On the SF-XL benchmark the method reports 18.3% and 6.7% gains over a prior classification-retrieval baseline on test sets v1 and v2 respectively, together with consistent plug-in improvements on representative VPR methods across SF-XL, MSLS and Pitts30k.

Significance. If the performance claims hold under controlled conditions, the work draws attention to an under-studied practical failure mode in large-scale urban VPR and supplies a lightweight, model-agnostic remedy. The explicit multi-benchmark evaluation and plug-in design are positive features that facilitate adoption.

major comments (2)
  1. [Abstract / Experiments] Abstract and Experiments section: the headline improvements (18.3% on SF-XL v1, 6.7% on v2) and the claim of consistent plug-in gains are presented without any description of experimental controls, error bars, statistical significance tests, or ablation studies that isolate the contribution of gradient rebalancing versus multi-scale search; this information is required to substantiate the central empirical claim.
  2. [Abstract] Abstract: the assertion that DAPR is model-agnostic and yields complementary gains without new biases rests on the premise that rebalancing weights, scale sets and compactness thresholds are held fixed across SF-XL, MSLS and Pitts30k; no statement confirms whether these hyperparameters were re-optimized per dataset, which directly affects the interpretation of the reported generalizability.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive feedback highlighting the need for stronger empirical substantiation and clearer statements on generalizability. We address each major comment below and will incorporate the requested clarifications and additional analyses in the revised manuscript.

read point-by-point responses

  1. Referee: [Abstract / Experiments] Abstract and Experiments section: the headline improvements (18.3% on SF-XL v1, 6.7% on v2) and the claim of consistent plug-in gains are presented without any description of experimental controls, error bars, statistical significance tests, or ablation studies that isolate the contribution of gradient rebalancing versus multi-scale search; this information is required to substantiate the central empirical claim.

    Authors: We agree that the current presentation lacks explicit documentation of experimental controls, error bars, statistical significance, and component-wise ablations. To address this, the revised manuscript will include: (i) ablation tables isolating gradient rebalancing from multi-scale distance search, (ii) error bars from multiple random seeds where computationally feasible, and (iii) statistical significance tests (e.g., paired t-tests) on the reported gains. These additions will be placed in the Experiments section with corresponding updates to the abstract. revision: yes

  2. Referee: [Abstract] Abstract: the assertion that DAPR is model-agnostic and yields complementary gains without new biases rests on the premise that rebalancing weights, scale sets and compactness thresholds are held fixed across SF-XL, MSLS and Pitts30k; no statement confirms whether these hyperparameters were re-optimized per dataset, which directly affects the interpretation of the reported generalizability.

    Authors: The hyperparameters (rebalancing weights, scale sets, and compactness thresholds) were held fixed across all three benchmarks precisely to demonstrate the plug-in, model-agnostic character of DAPR without per-dataset retuning. We will add an explicit statement to this effect in both the abstract and the method description in the revised manuscript to remove any ambiguity regarding generalizability. revision: yes

Circularity Check

0 steps flagged

No circularity in derivation chain

full rationale

The paper proposes an empirical plug-in framework (DAPR) for rebalancing gradients and multi-scale search in VPR, with all performance claims presented as measured outcomes on external benchmarks (SF-XL, MSLS, Pitts30k). No equations, derivations, or predictions are shown that reduce by construction to fitted inputs or self-citations; the central claims rest on benchmark results rather than any self-referential mechanism, satisfying the default expectation of a non-circular empirical contribution.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract provides no explicit free parameters, axioms, or invented entities; the method is described at conceptual level only.

pith-pipeline@v0.9.1-grok · 5749 in / 1124 out tokens · 23797 ms · 2026-07-02T19:53:26.214831+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

69 extracted references · 3 canonical work pages · 2 internal anchors

  1. [1]

    Neurocomputing513, 194–203 (2022)

    Ali-bey, A., Chaib-draa, B., Giguère, P.: Gsv-cities: Toward appropriate supervised visual place recognition. Neurocomputing513, 194–203 (2022)

    2022

  2. [2]

    In: Proceedings of the IEEE/CVF winter conference on applications of computer vision

    Ali-Bey, A., Chaib-Draa, B., Giguere, P.: Mixvpr: Feature mixing for visual place recognition. In: Proceedings of the IEEE/CVF winter conference on applications of computer vision. pp. 2998–3007 (2023)

    2023

  3. [3]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

    Ali-Bey, A., Chaib-draa, B., Giguère, P.: Boq: A place is worth a bag of learnable queries. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 17794–17803 (2024)

    2024

  4. [4]

    Anderson, C.: The long tail: why the future of business is selling less of more

  5. [5]

    In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (June 2016)

    Arandjelovic, R., Gronat, P., Torii, A., Pajdla, T., Sivic, J.: Netvlad: Cnn ar- chitecture for weakly supervised place recognition. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (June 2016)

    2016

  6. [6]

    In: Proceedings of the IEEE international conference on computer vision

    Babenko, A., Lempitsky, V.: Aggregating local deep features for image retrieval. In: Proceedings of the IEEE international conference on computer vision. pp. 1269– 1277 (2015)

    2015

  7. [7]

    Barbarani, G., Mostafa, M., Bayramov, H., Trivigno, G., Berton, G., Masone, C., Caputo, B.: Are local features all you need for cross-domain visual place recog- nition? In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 6155–6165 (2023)

    2023

  8. [8]

    In: Proceedings of the Computer Vision and Pattern Recognition Conference

    Berton, G., Masone, C.: Megaloc: One retrieval to place them all. In: Proceedings of the Computer Vision and Pattern Recognition Conference. pp. 2861–2867 (2025)

    2025

  9. [9]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

    Berton, G., Masone, C., Caputo, B.: Rethinking visual geo-localization for large- scale applications. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 4878–4888 (2022)

    2022

  10. [10]

    In: Proceedings of the IEEE/CVF Interna- tional Conference on Computer Vision

    Berton, G., Masone, C., Paolicelli, V., Caputo, B.: Viewpoint invariant dense matching for visual geolocalization. In: Proceedings of the IEEE/CVF Interna- tional Conference on Computer Vision. pp. 12169–12178 (2021)

    2021

  11. [11]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

    Berton, G., Mereu, R., Trivigno, G., Masone, C., Csurka, G., Sattler, T., Caputo, B.: Deep visual geo-localization benchmark. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 5396–5407 (2022)

    2022

  12. [12]

    In: Proceedings of the IEEE/CVF In- ternational Conference on Computer Vision

    Berton, G., Trivigno, G., Caputo, B., Masone, C.: Eigenplaces: Training viewpoint robust models for visual place recognition. In: Proceedings of the IEEE/CVF In- ternational Conference on Computer Vision. pp. 11080–11090 (2023)

    2023

  13. [13]

    In: ICASSP 2021-2021 IEEE International Confer- ence on Acoustics, Speech and Signal Processing (ICASSP)

    Chen, D., Chen, Y., Li, Y., Mao, F., He, Y., Xue, H.: Self-supervised learning for few-shot image classification. In: ICASSP 2021-2021 IEEE International Confer- ence on Acoustics, Speech and Signal Processing (ICASSP). pp. 1745–1749. IEEE (2021)

    2021

  14. [14]

    In: European conference on computer vision

    Chu, P., Bian, X., Liu, S., Ling, H.: Feature space augmentation for long-tailed data. In: European conference on computer vision. pp. 694–710. Springer (2020)

    2020

  15. [15]

    The annals of Statistics pp

    Feuerverger, A., Mureika, R.A.: The empirical characteristic function and its ap- plications. The annals of Statistics pp. 88–97 (1977)

    1977

  16. [16]

    In: European Conference on Computer Vision

    Garg, K., Puligilla, S.S., Kolathaya, S., Krishna, M., Garg, S.: Revisit anything: Visual place recognition via image segment retrieval. In: European Conference on Computer Vision. pp. 326–343. Springer (2024)

    2024

  17. [17]

    In: European conference on computer vision

    Ge, Y., Wang, H., Zhu, F., Zhao, R., Li, H.: Self-supervising fine-grained region similarities for large-scale image localization. In: European conference on computer vision. pp. 369–386. Springer (2020) Lost in the Tail: Geographic Imbalance in Urban VPR 17

    2020

  18. [18]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

    Haas, L., Skreta, M., Alberti, S., Finn, C.: Pigeon: Predicting image geolocations. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 12893–12902 (2024)

    2024

  19. [19]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

    Hausler, S., Garg, S., Xu, M., Milford, M., Fischer, T.: Patch-netvlad: Multi-scale fusion of locally-global descriptors for place recognition. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 14141– 14152 (2021)

    2021

  20. [20]

    IEEE Access (2024)

    Hu, J., Nie, J., Ning, Z., Feng, C., Wang, L., Li, J., Cheng, S.: Enhancing visual place recognition with hybrid attention mechanisms in mixvpr. IEEE Access (2024)

    2024

  21. [21]

    Hu, S., Feng, M., Nguyen, R.M., Lee, G.H.: Cvm-net: Cross-view matching net- workforimage-basedground-to-aerialgeo-localization.In:ProceedingsoftheIEEE Conference on Computer Vision and Pattern Recognition. pp. 7258–7267 (2018)

    2018

  22. [22]

    In: European Conference on Computer Vision

    Izquierdo, S., Civera, J.: Close, but not there: Boosting geographic distance sen- sitivity in visual place recognition. In: European Conference on Computer Vision. pp. 240–257. Springer (2024)

    2024

  23. [23]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

    Izquierdo,S.,Civera,J.:Optimaltransportaggregationforvisualplacerecognition. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). pp. 17658–17668 (June 2024)

    2024

  24. [24]

    In: Proceedings of the IEEE Conference on Computer Vi- sion and Pattern Recognition

    Jin Kim, H., Dunn, E., Frahm, J.M.: Learned contextual feature reweighting for image geo-localization. In: Proceedings of the IEEE Conference on Computer Vi- sion and Pattern Recognition. pp. 2136–2145 (2017)

    2017

  25. [25]

    In: 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition

    Jégou, H., Douze, M., Schmid, C., Pérez, P.: Aggregating local descriptors into a compact image representation. In: 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. pp. 3304–3311 (2010).https://doi. org/10.1109/CVPR.2010.5540039

    work page doi:10.1109/cvpr.2010.5540039 2010

  26. [26]

    In: Eighth In- ternational Conference on Learning Representations (ICLR) (2020)

    Kang, B., Xie, S., Rohrbach, M., Yan, Z., Gordo, A., Feng, J., Kalantidis, Y.: Decoupling representation and classifier for long-tailed recognition. In: Eighth In- ternational Conference on Learning Representations (ICLR) (2020)

    2020

  27. [27]

    IEEE Robotics and Automation Letters9(2), 1286–1293 (2023)

    Keetha, N., Mishra, A., Karhade, J., Jatavallabhula, K.M., Scherer, S., Krishna, M., Garg, S.: Anyloc: Towards universal visual place recognition. IEEE Robotics and Automation Letters9(2), 1286–1293 (2023)

    2023

  28. [28]

    In: European Conference on Computer Vision

    Khaliq, A., Xu, M., Hausler, S., Milford, M., Garg, S.: Vlad-buff: burst-aware fast feature aggregation for visual place recognition. In: European Conference on Computer Vision. pp. 447–466. Springer (2024)

    2024

  29. [29]

    Adam: A Method for Stochastic Optimization

    Kingma, D.P., Ba, J.: Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980 (2014)

    work page internal anchor Pith review Pith/arXiv arXiv 2014

  30. [30]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

    Li, M., Cheung, Y.m., Lu, Y.: Long-tailed visual recognition via gaussian clouded logit adjustment. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 6929–6938 (2022)

    2022

  31. [31]

    Lin, T.Y., Goyal, P., Girshick, R., He, K., Dollár, P.: Focal loss for dense object detection.In:ProceedingsoftheIEEEinternationalconferenceoncomputervision. pp. 2980–2988 (2017)

    2017

  32. [32]

    In: Proceedings of the IEEE/CVF International Conference on Computer Vision

    Liu, L., Li, H., Dai, Y.: Stochastic attraction-repulsion embedding for large scale image localization. In: Proceedings of the IEEE/CVF International Conference on Computer Vision. pp. 2570–2579 (2019)

    2019

  33. [33]

    Advances in Neural Information Processing Systems38, 85096–85124 (2026)

    Lu, F., Jin, T., Ye, C., Lan, X., Liu, Y., Yuan, C.: Towards implicit aggregation: Robust image representation for place recognition in the transformer era. Advances in Neural Information Processing Systems38, 85096–85124 (2026)

    2026

  34. [34]

    In: Proceed- 18 Z.Shu et al

    Lu, F., Lan, X., Zhang, L., Jiang, D., Wang, Y., Yuan, C.: Cricavpr: Cross-image correlation-aware representation learning for visual place recognition. In: Proceed- 18 Z.Shu et al. ings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 16772–16782 (2024)

    2024

  35. [35]

    In: The Twelfth Interna- tional Conference on Learning Representations (2024)

    Lu, F., Zhang, L., Lan, X., Dong, S., Wang, Y., Yuan, C.: Towards seamless adap- tation of pre-trained models for visual place recognition. In: The Twelfth Interna- tional Conference on Learning Representations (2024)

    2024

  36. [36]

    In: International Conference on Learning Represen- tations (2020)

    Menon, A.K., Jayasumana, S., Rawat, A.S., Jain, H., Veit, A., Kumar, S.: Long-tail learning via logit adjustment. In: International Conference on Learning Represen- tations (2020)

    2020

  37. [37]

    In: Proceedings of the European conference on computer vision (ECCV)

    Muller-Budack, E., Pustu-Iren, K., Ewerth, R.: Geolocation estimation of photos using a hierarchical model and scene classification. In: Proceedings of the European conference on computer vision (ECCV). pp. 563–579 (2018)

    2018

  38. [38]

    DINOv2: Learning Robust Visual Features without Supervision

    Oquab, M., Darcet, T., Moutakanni, T., Vo, H., Szafraniec, M., Khalidov, V., Fernandez, P., Haziza, D., Massa, F., El-Nouby, A., et al.: Dinov2: Learning robust visual features without supervision. arXiv preprint arXiv:2304.07193 (2023)

    work page internal anchor Pith review Pith/arXiv arXiv 2023

  39. [39]

    In: International Conference on Image Analysis and Processing

    Paolicelli, V., Tavera, A., Masone, C., Berton, G., Caputo, B.: Learning seman- tics for visual place recognition through multi-scale attention. In: International Conference on Image Analysis and Processing. pp. 454–466. Springer (2022)

    2022

  40. [40]

    In: Proceedings of the IEEE/CVF International Conference on Computer Vision

    Peng, G., Zhang, J., Li, H., Wang, D.: Attentional pyramid pooling of salient visual residuals for place recognition. In: Proceedings of the IEEE/CVF International Conference on Computer Vision. pp. 885–894 (2021)

    2021

  41. [41]

    IEEE transactions on pattern analysis and machine intelligence 41(7), 1655–1668 (2018)

    Radenović, F., Tolias, G., Chum, O.: Fine-tuning cnn image retrieval with no human annotation. IEEE transactions on pattern analysis and machine intelligence 41(7), 1655–1668 (2018)

    2018

  42. [42]

    In: Proceedings of the European Conference on Computer Vision (ECCV)

    Seo, P.H., Weyand, T., Sim, J., Han, B.: Cplanet: Enhancing image geolocalization by combinatorial partitioning of maps. In: Proceedings of the European Conference on Computer Vision (ECCV). pp. 536–551 (2018)

    2018

  43. [43]

    In: Proceedings of the Computer Vision and Pattern Recognition Conference

    Sferrazza, D., Berton, G., Trivigno, G., Masone, C.: To match or not to match: Revisiting image matching for reliable visual place recognition. In: Proceedings of the Computer Vision and Pattern Recognition Conference. pp. 2849–2860 (2025)

    2025

  44. [44]

    In: Proceedings of the IEEE/CVF International Conference on Computer Vision

    Shao, S., Chen, K., Karpur, A., Cui, Q., Araujo, A., Cao, B.: Global features are all you need for image retrieval and reranking. In: Proceedings of the IEEE/CVF International Conference on Computer Vision. pp. 11036–11046 (2023)

    2023

  45. [45]

    Shi, J.X., Wei, T., Xiang, Y., Li, Y.F.: How re-sampling helps for long-tail learning? Advances in Neural Information Processing Systems36, 75669–75687 (2023)

    2023

  46. [46]

    In: Workshop on Long-Term Auton- omy, IEEE International Conference on Robotics and Automation (ICRA) (2013)

    Sünderhauf, N., Neubert, P., Protzel, P.: Are we there yet? challenging SeqSLAM on a 3000km journey across all four seasons. In: Workshop on Long-Term Auton- omy, IEEE International Conference on Robotics and Automation (ICRA) (2013)

    2013

  47. [47]

    In: Proceedings of the IEEE conference on computer vision and pattern recognition

    Torii, A., Arandjelovic, R., Sivic, J., Okutomi, M., Pajdla, T.: 24/7 place recogni- tion by view synthesis. In: Proceedings of the IEEE conference on computer vision and pattern recognition. pp. 1808–1817 (2015)

    2015

  48. [48]

    In: Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV)

    Trivigno, G., Berton, G., Aragon, J., Caputo, B., Masone, C.: Divide&classify: Fine-grained classification for city-wide visual geo-localization. In: Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV). pp. 11142– 11152 (October 2023)

    2023

  49. [49]

    In: International Conference on Learning Representations

    Tzachor, I., Lerner, B., Levy, M., Green, M., Berkovitz Shalev, T., Habib, G., Samuel, D., Zailer, N., Shimshi, O., Darshan, N., et al.: Effovpr: Effective founda- tion model utilization for visual place recognition. In: International Conference on Learning Representations. vol. 2025, pp. 42817–42839 (2025)

    2025

  50. [50]

    In: The Twelfth International Conference on Learning Representations (2024), https://openreview.net/forum?id=RzY9qQHUXy

    Wang, B., Wang, P., Xu, W., Wang, X., Zhang, Y., Wang, K., Wang, Y.: Kill two birds with one stone: Rethinking data augmentation for deep long-tailed learn- Lost in the Tail: Geographic Imbalance in Urban VPR 19 ing. In: The Twelfth International Conference on Learning Representations (2024), https://openreview.net/forum?id=RzY9qQHUXy

    2024

  51. [51]

    In: Proceedings of the IEEE conference on computer vision and pattern recognition

    Wang, H., Wang, Y., Zhou, Z., Ji, X., Gong, D., Zhou, J., Li, Z., Liu, W.: Cosface: Large margin cosine loss for deep face recognition. In: Proceedings of the IEEE conference on computer vision and pattern recognition. pp. 5265–5274 (2018)

    2018

  52. [52]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

    Wang, R., Shen, Y., Zuo, W., Zhou, S., Zheng, N.: Transvpr: Transformer-based place recognition with multi-level attention aggregation. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 13648– 13657 (2022)

    2022

  53. [53]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

    Wang, S., Yang, Y., Liu, Z., Sun, C., Hu, X., He, C., Zhang, L.: Dataset distillation with neural characteristic function: A minmax perspective. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). pp. 25570–25580 (June 2025)

    2025

  54. [54]

    Advances in Neural Information Processing Systems36, 48417–48430 (2023)

    Wang, Z., Xu, Q., Yang, Z., He, Y., Cao, X., Huang, Q.: A unified generalization analysis of re-weighting and logit-adjustment for imbalanced learning. Advances in Neural Information Processing Systems36, 48417–48430 (2023)

    2023

  55. [55]

    IEEE Transactions on Pattern Analysis and Machine Intelligence (2025)

    Wang, Z., Xu, Q., Yang, Z., Xu, Z., Zhang, L., Cao, X., Huang, Q.: A unified per- spective for loss-oriented imbalanced learning via localization. IEEE Transactions on Pattern Analysis and Machine Intelligence (2025)

    2025

  56. [56]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recog- nition

    Warburg, F., Hauberg, S., Lopez-Antequera, M., Gargallo, P., Kuang, Y., Civera, J.: Mapillary street-level sequences: A dataset for lifelong place recognition. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recog- nition. pp. 2626–2635 (2020)

    2020

  57. [57]

    In: Proceedings of the IEEE/CVF Inter- national Conference on Computer Vision (ICCV)

    Warburg, F., Jørgensen, M., Civera, J., Hauberg, S.: Bayesian triplet loss: Uncer- tainty quantification in image retrieval. In: Proceedings of the IEEE/CVF Inter- national Conference on Computer Vision (ICCV). pp. 12158–12168 (2021)

    2021

  58. [58]

    In: European conference on computer vision

    Weyand, T., Kostrikov, I., Philbin, J.: Planet-photo geolocation with convolutional neural networks. In: European conference on computer vision. pp. 37–55. Springer (2016)

    2016

  59. [59]

    In: European conference on computer vision

    Wu, T., Huang, Q., Liu, Z., Wang, Y., Lin, D.: Distribution-balanced loss for multi- label classification in long-tailed datasets. In: European conference on computer vision. pp. 162–178. Springer (2020)

    2020

  60. [60]

    In: Forty-first International Conference on Machine Learning (2024)

    Yang, Z., Xu, Q., Wang, Z., Li, S., Han, B., Bao, S., Cao, X., Huang, Q.: Harnessing hierarchical label distribution variations in test agnostic long-tail recognition. In: Forty-first International Conference on Machine Learning (2024)

    2024

  61. [61]

    ACM Transactions on Human-Robot Interaction14(2), 1–36 (2024)

    Ye, X., Robert, L.P.: A human–security robot interaction literature review. ACM Transactions on Human-Robot Interaction14(2), 1–36 (2024)

    2024

  62. [62]

    Pattern Recognition161, 111200 (2025)

    Yu, H., Du, Y., Wu, J.: Reviving undersampling for long-tailed learning. Pattern Recognition161, 111200 (2025)

    2025

  63. [63]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

    Zaffar, M., Nan, L., Kooij, J.F.P.: On the estimation of image-matching uncer- tainty in visual place recognition. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). pp. 17743–17753 (2024)

    2024

  64. [64]

    In: Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision

    Zhang, H., Chen, X., Jing, H., Zheng, Y., Wu, Y., Jin, C.: Etr: An efficient trans- former for re-ranking in visual place recognition. In: Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision. pp. 5665–5674 (2023)

    2023

  65. [65]

    In: Proceedings of the IEEE/CVF International Conference on Computer Vision

    Zhang, S., Mao, H., Chen, Q., Kim, Y.: Efficient visual place recognition through multimodal semantic knowledge integration. In: Proceedings of the IEEE/CVF International Conference on Computer Vision. pp. 5601–5610 (2025)

    2025

  66. [66]

    IEEE transactions on pattern analysis and machine intelligence45(9), 10795– 10816 (2023) 20 Z.Shu et al

    Zhang, Y., Kang, B., Hooi, B., Yan, S., Feng, J.: Deep long-tailed learning: A sur- vey. IEEE transactions on pattern analysis and machine intelligence45(9), 10795– 10816 (2023) 20 Z.Shu et al

    2023

  67. [67]

    In: Proceedings of the AAAI conference on artificial intelligence

    Zhao, Y., Chen, W., Tan, X., Huang, K., Zhu, J.: Adaptive logit adjustment loss for long-tailed visual recognition. In: Proceedings of the AAAI conference on artificial intelligence. vol. 36, pp. 3472–3480 (2022)

    2022

  68. [68]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

    Zhu, S., Yang, L., Chen, C., Shah, M., Shen, X., Wang, H.: R2former: Unified retrieval and reranking transformer for place recognition. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 19370– 19380 (2023)

    2023

  69. [69]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

    Zhu, S., Yang, T., Chen, C.: Vigor: Cross-view image geo-localization beyond one- to-one retrieval. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 3640–3649 (2021) Lost in the Tail: Geographic Imbalance in Urban VPR 1 Lost in the Tail: Addressing Geographic Imbalance in Urban Visual Place Recognition Supplementa...

    2021